Structural use of bamboo: Part 4:

Element design equations

Laurence, AT, Trujillo, D, Feltham, I, Kaminski, S & Delgado, FL
Author post-print (accepted) deposited by Coventry University’s Repository

Original citation & hyperlink:

Laurence, AT, Trujillo, D, Feltham, I, Kaminski, S & Delgado, FL 2017, 'Structural use
of bamboo.: Part 4: Element design equations' The Structural Engineer, vol 95, no. 3,
pp. 24-27
https://www.istructe.org/journal/volumes/volume-95-(2017)/issue-3/structural-
use-of-bamboo-part-4-element-design-e

ISSN 1466-5123

Publisher: Institution of Structural Engineers

Copyright © and Moral Rights are retained by the author(s) and/ or other copyright
owners. A copy can be downloaded for personal non-commercial research or study,
without prior permission or charge. This item cannot be reproduced or quoted extensively
from without first obtaining permission in writing from the copyright holder(s). The
content must not be changed in any way or sold commercially in any format or medium
without the formal permission of the copyright holders.

This document is the author’s post-print version, incorporating any revisions agreed during
the peer-review process. Some differences between the published version and this version
may remain and you are advised to consult the published version if you wish to cite from
it.
Technical Note Series: Structural Use of Bamboo

Technical Note 4: Element design equations

Sebastian Kaminski, MEng (Hons) ACGI CEng MIStructE, Senior Structural Engineer at
Arup Advanced Technology + Research London, Member of INBAR Task Force – Bamboo
Construction

Andrew Lawrence, MA (Cantab) CEng MICE MIStructE, Associate Director at Arup

Advanced Technology + Research London, Member of INBAR Task Force – Bamboo
Construction

David Trujillo, MSc DIC CEng MIStructE, Senior Lecturer at Coventry University, Chair of
INBAR Task Force – Bamboo Construction

Ian Feltham, MA CEng MICE MIStructE, Director and Fellow at Arup Advanced
Technology + Research London,

Luis Felipe López MSc CEng MIStructE, Head of Engineering Department at Base Bahay
Foundation the Philippines, Member of Colombian Earthquake Engineering Association
(AIS)

Synopsis

Bamboo is a strong, fast growing and very sustainable material, having been used structurally
for thousands of years in many parts of the world. In modern times, it has the potential to be
an aesthetically pleasing and cost-competitive alternative to more conventional materials,
such as timber, as demonstrated by some visually impressive recent structures.

This five-part technical series, aimed at both developed- and developing-world contexts, will
bring together current knowledge and best practice on the structural use of bamboo, covering:

 an introduction to bamboo (part 1)

 durability and preservation (part 2)
 design values (part 3)
 element design equations (part 4)
 connections (part 5)

This fourth article proposes element design equations for typical loading conditions on a
bamboo culm: flexure, shear, axial tension, axial compression and local bearing. The work
draws on existing published bamboo and timber design codes, and sound engineering
judgement.

Introduction
Element design in bamboo is relatively similar to timber and must always follow elastic stress
distributions since bamboo is a brittle material. The methods proposed are based on a Limit
State approach and have been developed based on ISO 22156: Bamboo – Structural Design1,
ISO 22157: Bamboo – Determination of Physical and Mechanical Properties2,3, NSR-10 G-
12: Colombian Code for Seismically-Resistant Construction: Structures of Timber and
Guadua Bamboo4, BS5268-2: Structural use of timber5 and EN 1995-1-1: Eurocode 5:
Design of Timber Structures6. Design strengths should be taken from Structural use of
bamboo: Part 3: Design values7.

When determining loads in bamboo elements, an approach similar to that used in timber
structures should be used – i.e. a view should be taken on how the load path might vary
depending on the possible variations in stiffness of the connections and elements. This is
particularly important because most failure modes in bamboo elements and bolted
connections are brittle. This is best done by a sensitivity analysis of the structural load path,
using good engineering judgement.

For all elements and all structures, it is important to use bamboo that has been dried to a
moisture content close to the moisture content of equilibrium in the location where the
structure will be built. Failure to do so may result in splits in elements and joints that may
compromise the strength of the structure.

It is recommended that for safety critical elements, such as primary beams and columns,
designers consider bundling culms together. Although composite action flexurally is unlikely
to be able to be justified and therefore cannot be exploited, bundling has the potential to make
the overall element less sensitive to splitting of individual culms, hence they are more robust.
Similarly, consideration should be given in design to the possibility of replacing individual
elements in service should these exhibit severe splitting – bundling may aid this process due
to redundancy in the element.

Nomenclature

Nomenclature used throughout this Note:

bundle = several culms bundled together forming an element, generally sharing load but not
acting compositely
culm = stem or stalk of bamboo
element = a part of the structure (beam, column etc.), formed from one or more culms
fibres = the cellulose fibres of the bamboo which run longitudinally along the length of the
culm
internode = the culm region between nodes
lignin = a weak matrix which holds the longitudinal fibres of the culm together
node = a solid band at regular intervals along the culm, where the fibres acts as a diaphragm
and close off the section
splits = a longitudinal crack along the length of the culm

b = length of loaded area for local crushing (mm)

𝑑𝑖 = distance between the centroid of the combined section and the centroid of each
individual culm (mm)
kcr = crack factor
𝑋𝑖,𝑑 = design strength (N/mm2)7
𝑋𝑐,0,𝑑 = design compressive strength parallel to fibre (N/mm2)7
𝑋𝑚,𝑑 = design flexural strength (N/mm2)7
𝑋𝑡,0,𝑑 = design tensile strength parallel to fibre (N/mm2)7
𝑋𝑣,𝑑 = design shear strength about primary axis about either axis (N/mm2)7
𝑘 = coefficient of effective length
𝑙𝑒 = effective length of element (mm)
𝑟 = radius of gyration of section/composite section
t = average wall thickness of culm, based on the average of the eight dimensions shown in
Figure 1 (four at each end) (mm)
𝛾𝐸 = material factor of safety for modulus of elasticity
𝜆 = slenderness of element/composite element

A = net area of section (less any holes) (mm2)

𝐴𝑖 = net section area (less any holes) of each individual culm (mm2)
A𝑡𝑜𝑡 = sum of the net section areas (less any holes) of the individual culms that make up the
compressive element (mm2)
𝐶𝑘 = boundary coefficient between intermediate and long slender elements for axial
compression
De = average external diameter of culm (mm), based on the average of the four diameters
shown in Figure 1 (two at each end)
𝐷𝑜𝑢𝑡 = external diameter of second culm when checking local crushing (mm)
𝐸0.05 = modulus of elasticity, 5th percentile (N/mm2)
𝐹𝑖 = design capacity of element (N)
𝐹𝑐 = design compressive capacity of element (N)
𝑀𝑚 = design flexural capacity of element (Nmm)
𝐹𝑡 = design tensile capacity of element (N)
𝐹𝑣 = design shear capacity of element (N)
𝐼 = second moment of area of combined/composite section (mm4)
𝐼𝑖 = second moment of area of each individual culm forming the section (mm4)
L = length between points at which the element is restrained against buckling in that plane
(mm)
𝑀 = moment applied to element (Nmm)
𝑁𝑐 = axial compressive force applied to element (N)
𝑁𝑒 = Euler critical load for axial compression (N)
𝑁𝑡 = axial tensile force applied to element (N)
𝑆𝑒𝑙𝑎𝑠𝑡𝑖𝑐 = elastic section modulus (mm3)
𝑉 = shear force applied to element (N)

1.0 Dimensions for structural design

Culm dimensions and basic properties can be determined as follows and from Figure 1:

Figure 1: Locations of external diameters and wall thicknesses to measure (refer to

nomenclature).

The net area of section, A (less any bolt holes), can be determined by the following equation:

𝜋
𝐴 = [ 4 (𝐷𝑒 2 − (𝐷𝑒 − 2𝑡)2 )] − ∑ 𝑎𝑟𝑒𝑎 𝑜𝑓 𝑎𝑛𝑦 ℎ𝑜𝑙𝑒𝑠 Eq 1
2.0 Element failure modes
2.1 Flexure
The capacity of a bamboo culm in flexure is limited by the transverse strain capacity of the
lignin8 (which is very roughly 1.1×10-3 – see Technical Note 19), hence flexural failure
generally manifests itself as a collapse by separation of the fibres (similar to bursting or local
buckling) in the compressive zone10.

The design flexural capacity of an element, Mm, can be determined assuming an elastic stress
distribution and based on the average wall thicknesses and diameters determined in Section
Error! Reference source not found.. Local buckling of walls cannot be ruled out. For
species where the De/t ratio is larger than 11, this possibility needs to be examined either
through analysis or experimentation.

𝑀𝑚 = 𝑋𝑚,𝑑 𝑆𝑒𝑙𝑎𝑠𝑡𝑖𝑐 Eq 2

where:

𝜋(𝐷𝑒 4 −[𝐷𝑒 −2𝑡]4 )

𝑆𝑒𝑙𝑎𝑠𝑡𝑖𝑐 = Eq 3
32𝐷𝑒

Note that to date a connection system that allows composite action between culms in bending
has not been developed – for further details refer to Trujillo and Archila11.

Deflection of beams in bending often governs above flexure and should therefore normally be
checked – see Section 4.0.

2.2 Shear
The capacity of a bamboo culm in shear is limited by the longitudinal shear capacity of the
lignin between the fibres.

The design shear capacity of an element, Fv, can be determined assuming an elastic
(approximately sinusoidally) stress distribution utilising the entire section of the culm, and
based on the average wall thickness and diameter.

3𝜋𝑡(𝐷𝑒 4 −(𝐷𝑒 −2𝑡)4 )

𝐹𝑣 = 𝑋𝑣,𝑑 𝑘𝑐𝑟 Eq 4
8(𝐷𝑒 3 −(𝐷𝑒 −2𝑡)3 )

Longitudinal splits are common in bamboo (Figure 2) and are difficult to avoid, therefore it is
recommended to apply a crack reduction factor, kcr, similar to that used in EN 1995-1-16. The
recommended value for kcr is given as follows:

𝑘𝑐𝑟 = 0.5

This factor accounts for the risk of a single split through the culm wall.
Figure 2: Splits in bamboo

2.3 Axial tension

The capacity of a bamboo culm in axial tension is limited by the tensile capacity of the fibres,
however, this rarely governs since it is very difficult to mobilise the full axial tension
capacity of a culm because the connections will tend to fail first in other mechanisms such as
shear, local crushing or tension perpendicular to fibre.

The design allowable tensile capacity of an element, Ft, can be determined utilising the net
area of the section (less any holes), and based on the average wall thickness and diameter.

𝐹𝑇 = 𝑋𝑡,0,𝑑 𝐴 Eq 5

2.4 Axial compression

The capacity of a bamboo culm in axial compression is limited by: local crushing of the
fibres in short elements (~L<2De), bursting of the culm wall and separation of the fibres (due
to failure of the lignin holding the fibres together) in medium length elements (~L>2De), and
global Euler buckling of the section in long elements (~L>10De). The design allowable axial
compressive capacity of an element, Fc, can be determined as per the following section. The
approach uses equations which follow a Perry-Robertson curve with an initial imperfection of
3%, which assumes a maximum out-of-straightness of 2% (i.e., no part of the culm should
deviate more than 20mm per metre length from a straight line drawn between the ends of a
culm); the other 1% of the imperfection is a notional allowance for variations in the geometry
of the cross section and the material properties. The approach uses equations which follow a
Perry-Robertson curve with an initial imperfection of 3%, which assumes a maximum out-of-
straightness of 1% (i.e., drawing a straight line between the ends of the culms, no part of the
culm should deviate more than 10mm per metre length outside this). This out-of-straightness
limit would need to be part of any grading regime. For some species of bamboo this may be
difficult to achieve: where the out-of-straightness exceeds 12%, the additional moment
induced by this PΔ effect with the axial load should be considered together with the axial
load.

Local buckling of walls cannot be ruled out. For species where the De/t ratio is larger than 11,
this possibility needs to be examined either through analysis or experimentation.

2.4.1 Effective length, le

𝑙𝑒 = 𝑘𝐿 Eq 6

where k is taken from Figure 2.

 Support Non-sway mode, effectively held in position at both Sway mode, effectively held in position and restrained in
conditions ends, not restrained in direction at either end direction at one end, not held in position or restrained in
direction at other end
k 1.0 2.1
Figure 2: Coefficient of effective length, k

Note that other typical coefficients of effective length can theoretically be used (based on
achieving a distance between points of contraflexure). However, due to the difficulty in
actively restraining bamboo culms against rotation, the limited number of fasteners that can
be used in connections and the brittle nature of connections, consideration should be given to
whether connections can provide adequate restraint.

2.4.2 Slenderness, 𝜆
𝑙
𝜆 = 𝑟𝑒 Eq 7

2.4.3 Net area of section, Atot

𝐴𝑡𝑜𝑡 = ∑ 𝐴𝑖 Eq 8

2.4.4 Radius of gyration, r

0.9𝐼
𝑟 = √𝐴 Eq 9
𝑡𝑜𝑡

The second moment of area has been reduced by 10% to account for the effect of the taper of
bamboo1.

2.4.5 Second moment of area, I

2.4.5.1 Elements formed from a single culm
For an element formed from a single culm:

(𝐷𝑒 4 −[𝐷𝑒 −2𝑡]4 )

𝐼=𝜋 Eq 10
64

2.4.5.2 Elements formed from multiple culms

It may be possible to achieve composite action in columns formed by multiple culms bundled
together, however, to the authors’ knowledge this has not been demonstrated
experimentally12.

Therefore whenever an element is formed from multiple culms, and composite action is not
demonstrable, the second moment of area (or moment of inertia) for the element would be
calculated thus:
𝐼 = ∑ 𝐼𝑖 Eq 11

An element formed from multiple culms can be considered to act compositely in axial
compression if full scale experimentation demonstrates that the experimentally measured
second moment resembles that of Eq 12.

𝐼𝑡𝑒𝑠𝑡 ≈ 𝐼𝑐𝑜𝑚𝑝𝑜𝑠𝑖𝑡𝑒 = ∑(𝐴𝑖 ∙ 𝑑𝑖2 ) + ∑ 𝐼𝑖 Eq. 12

2.4.6 Slenderness classification

Table 1 can be used to classify the element according to its slenderness from cl. 2.4.2 and Eq
7.

Table 1: Column slenderness classification (from NSR-10 G-124)

Column classification Slenderness
Short (stocky) λ < 30
Intermediate (slender) 30 < λ < Ck
Long (very slender) Ck < λ < 150

where (from NSR-10 G-124):

𝐸0.05
𝐶𝑘 = 𝜋√𝛾 Eq 13
𝐸 𝑋𝑐,0,𝑑

The γE is a material factor of safety, and its recommended value is given as follows:

𝛾𝐸 = 1.5

In no circumstances should an element have a slenderness λ > 150.

2.4.7 Capacity
2.4.7.1 Short (stocky) elements, λ < 30
For stocky elements, the compressive capacity can be based on the squash load.

𝐹𝑐 = 𝑋𝑐,0,𝑑 𝐴𝑡𝑜𝑡 Eq 14

2.4.7.2 Intermediate (slender) elements, 30 < λ < Ck

For intermediate slender elements, the compressive capacity can be a transition curve (from
NSR-10 G-124).

2 𝜆 3
𝐹𝑐 = 𝐴𝑡𝑜𝑡 𝑋𝑐,0,𝑑 (1 − 5 [𝐶 ] ) Eq 15
𝑘

2.4.7.3 Long (very slender) elements, Ck < λ < 150

For long, very slender elements, the compressive capacity can be based on Euler.

𝜋 2 𝐴𝑡𝑜𝑡 𝐸0.05
𝐹𝑐 = 𝑁𝑒 = Eq 16
𝛾𝐸 𝜆2
The γE is a material factor of safety, and its recommended value is given as follows:

𝛾𝐸 = 1.5

2.5 Elements subject to combined bending and axial tension

For elements subject to combined bending and axial tension, the sum of the utilisations
should be ≤ 1:

𝑁𝑡 M
+ MF ≤ 1 Eq 17
Ft m

2.6 Elements subject to combined bending and axial compression

For elements subject to combined bending and axial compression, the sum of the utilisations
should be ≤ 1:

𝑁𝑐 M
+ 1.5Nc ≤1 Eq 18
Fc MFm (1− )
Ne

where

π2 E0.05 I
Ne = Eq 19
le 2

3.0 Local failure modes: bearing (crushing)

Bamboo is particularly susceptible to crushing under point loads and over supports. Without
infilling the internode with mortar, crushing occurs by splitting of the sides of the culm
outwards, and splitting of the top and bottom of the culm inwards. Where the bamboo is
carrying a significant load (in practice in most cases), the internode should be infilled with an
iuncompressible material such as cement mortar and the capacity should be checked as will
be outlined in Article 5. Where the bamboo is not carrying a significant load (e.g., roof beams
supporting a light roof with no live load) infilling the internode may be avoided, the
procedure for checking the capacity will also be outlined in Article 5.

4.0 Deflection
Deflection checks of bamboo elements in bending under service loads often govern above
flexural capacity. These can be checked using standard beam theory, using I as defined in
Equation 10. Ranges of moduli of elasticity are suggested in Article 37, and should be
selected based on the confidence required in the deflection calculation.

When checking deflections of a whole structure, consider accounting for slip in connections,
especially if nailed or bolted.

Summary
This article proposes element design equations for typical loading conditions on a bamboo
culm: flexure, shear, axial tension, axial compression and local bearing. Structural design of
bamboo structures should follow a similar approach to that used in timber structures should
be used – i.e. a view should be taken on how the load path might vary depending on the
possible variations in stiffness of the connections and elements. Element design should
always follow elastic stress distributions, and existing defects (such as cracks/splits), taper
and out-of-straightness need to be considered – these can be controlled through a robust
selection regime. It is important to use bamboo that has been dried to a moisture content close
to the equilibrium moisture content in service of equilibrium in the location where the
structure will be built.

Areas of research that would be particularly beneficial with regards to element design are
composite action between culms in flexure and composite action between culms working as a
slender bunched column under axial compression – to-date these have been very difficult to
achieve in the laboratory.

An update to the ISO bamboo standards1,2,3 is currently underway and due for publication
within the next few years. These standards should incorporate the state-of-the-art in bamboo
research, and therefore are recommended to be used for bamboo design once available.

The next article in the series will cover connections.

References

1. International Standards Office (ISO) (2004) ISO22156. Bamboo: Structural design.

Geneva, ISO
2. International Standards Office (ISO) (2004) ISO22157-1. Bamboo: Determination of
physical and mechanical properties. Part 1: Requirements. Geneva, ISO
3. International Standards Office (ISO) (2004) ISO22157-2. Bamboo: Determination of
physical and mechanical properties. Part 2: Laboratory manual. Geneva, IS.
4. Asociación Colombiana de Ingeniería Sísmica (AIS) (2010) NSR-10: Reglamento
Colombiano de construcción sismo resistente. Titulo G: Estructuras de madera y
estructuras de guadua. Bogota, AIS.
5. British Standards Institute (BSI) (1999) BS 5268-2:1996: Structural use of Timber –
Part 2: Code of practice for permissible stress design, materials and workmanship
6. BS EN 1995-1-1 (2004+A1) Eurocode 5: Design of Timber Structures – Part 1-1:
General – Common rules and rules for buildings (CEN, 2009)
7. Kaminski S, Lawrence A, Trujillo D, Feltham I & López L (2016) ‘Structural Use of
Bamboo 3: Design values’. In The Structural Engineer, December 2016, pp. 42-45
8. Janssen J (2000) INBAR Technical Report 20: Designing and Building with Bamboo.
Beijing: INBAR
9. Kaminski S, Lawrence A, Trujillo D (2016) Structural use of bamboo: Part 1:
Introduction to bamboo. In The Structural Engineer, Aug 2016, pp. 40-43
10. Trujillo D, Jangra S & Gibson J (2016) Flexural properties as a basis for bamboo
strength grading. In Structures and Building – Proceedings of the Institution of Civil
Engineers The institution of Civil Engineers, London (DOI: 10.1680/jstbu.16.00084)
11. Trujillo D & Archila H (2016) Engineered bamboo and bamboo engineering –
TRADA Research Summary, Exova BM TRADA, High Wycombe. ISBN: 978-1-
909594-56-2
12. Richard M. & Harries K. (2012) Experimental Buckling Capacity of Multiple-Culm
Bamboo Columns. In Key engineering materials (Vol. 517, pp. 51-62). Trans Tech
Publications.